Think Piece: Good and bad arguments against positive discrimination

The US Supreme Court has just left one Texan affirmative action scheme in place, but it has recently busted schemes elsewhere. I discuss what libertarians should think about positive discrimination and affirmative action.

Many of the arguments libertarians make against affirmative action/positive discrimination do not hold. For example, it neither needs to interfere with equality before the law, nor does it need to imposed by state coercion. And in its favour, affirmative action may be one way to overcome some of unjust forms on inequality in our society. On the other hand, it is clearly not even close to the best way of dealing with unjust inequality. And some evidence suggests that these schemes actually hurt those they are designed to help. But without sufficient evidence perhaps the best short-term approach is to allow universities to experiment with their admissions process, so they can among them discover the best approach.

Read the whole thing.

Good and bad objections to positive discrimination

The US Supreme Court has just left one Texan affirmative action scheme in place, but it has recently busted schemes elsewhere. I discuss what libertarians should think about positive discrimination and affirmative action.

Many of the arguments libertarians make against affirmative action/positive discrimination do not hold. For example, it neither needs to interfere with equality before the law, nor does it need to imposed by state coercion. And in its favour, affirmative action may be one way to overcome some of unjust forms on inequality in our society. On the other hand, it is clearly not even close to the best way of dealing with unjust inequality. And some evidence suggests that these schemes actually hurt those they are designed to help. But without sufficient evidence perhaps the best short-term approach is to allow universities to experiment with their admissions process, so they can among them discover the best approach.

The supreme court on 24th June effectively decided to leave a Texan affirmative action programme in place. The programme had two main provisions: (1) guaranteeing any pupil finishing in the top 10% of their year group a place at a publicly-funded university in the Lone Star state; and (2) allowing university administrators to consider race and diversity as part of their admissions criteria.

Both elements work as positive discrimination tools. Guaranteeing places to relatively—as opposed to absolutely—high-achieving pupils should mean that at least some of the effects of school quality are subtracted out. And allowing authorities to take into account race means that deprivation or oppression experienced mainly or exclusively by non-whites may also be factored out. How do we judge whether this is bad or good?

Many libertarians would object that a system of affirmative action either because it violates equality before the law or because it interfered with people's free interactions, and could only come from state involvement. But neither of these claims necessarily hold.

Though it may not necessarily be true in the specifics of this US case, in general we'd generally expect and require that equal application of law took circumstances into account. For example, years of domestic abuse would rightly be considered an extenuating circumstance in a case where a parter accidentally went too far in self defence. Similarly equality before the law in schooling may require taking into account the statistically likely backgrounds of applicants.

The alternative seems less equal, since getting top A-levels at a tough inner-city comprehensive is surely more difficult than at a highly selective school, even given similar parental support. And in still-discriminatory societies people of colour usually have more difficult lives, even past their general social deprivation, and thus we might expect lower grades, even for an equally talented or conscientious student.

The other libertarian objection may have slightly more force. It would be rational for universities that either sought to maximise wealth (through boosting bequests) or academic prestige (through the best students or best research) to neuter out factors that affected school-level performance but would not affect university or later career performance. The factors listed seem like obvious examples of these. But affirmative action may have to go further, not simply aiming for top potential, but also for those who due to their unfairly poor circumstances have lower potential. It would seem to have to go this far if it wanted to attain the goal of fully accounting for circumstances, as some circumstances may reduce potential as well as reducing results in lower tiers of education. In effect, with affirmative action we want to act as if people had never been hampered with worse starts, even if at the university level we want to "leave in" differences in natural talents.

Thus it might be that state pressure would be necessary to get universities to consider more than just student potential. But this is far from certain; the University of Michigan's scheme, which gave 20 automatic points (out of an 100 needed for guaranteed admission) to underrepresented ethnic minorities. It was ruled unconstitutional but while it was in place it was chosen freely by the university. If chosen upon freely then surely libertarians should laud the schemes as admirable voluntary attempts at distributive equality through free association.

But there's a more telling objection to affirmative action: it causes more harm than it does good. In his barnstorming dissent to the judgement, (black) conservative justice Clarence Thomas points out many of the bad effects of the scheme. In his 2007 memoir he said "As much as it stung to be told that I'd done well in the seminary DESPITE my race, it was far worse to feel that I was now at Yale BECAUSE of it." And in his very readable judgement, along with pointing out the similarity—despite their apparently inverse goals—between segregationist and pro-slavery arguments and pro-affirmative action cases, he listed the negative impacts positive discrimination can have on those it's supposed to help.

Thomas cites a 2003 book which claims that "it is a fact that in virtually all selective schools…where racial preferences in admission is practiced, the majority of [black] students end up in the lower quarter of their class." And, according to Thomas, this upward shifting does not result in higher proportions of black or Hispanic students in higher education on average. Instead, minority students go to more selective schools than they would have otherwise attended, which he believes explains their relatively poor performances. This, in turn, pushes them into less challenging schools within those universities, he says, citing figures showing disproportionately high numbers of blacks and Hispanics study social work or education.

Worst of all for Thomas is that—in our flawed society (which won't be changed by this policy alone)—positive discrimination "stamps blacks and Hispanics with a badge of inferiority". One feature of race is that it's often outwardly identifiable, meaning any black or Hispanic student could be seen by others as owing their place to affirmative action, even though most of them in the particular case in question did not gain their place due to the system. He quotes a black student: "I was never able to be as proud of getting into Stanford as my classmates could be…how much of an achievement can I truly say it was to have been a good enough black person to be admitted, while my colleagues had been considered good enough people to be admitted?" This ties into the ideas Elizabeth Anderson explores in her classic essay "What is the Point of Equality?". She imagines a government grant to the less attractive—would they be grateful for the money or would they be deeply hurt by the elevation of subjective preferences to official dictum?

Such concerns, especially packaged with the litany of practical issues and inconsistencies highlighted by Thomas and Prof. Mark J Perry make the issue more difficult, and give us reason to question whether the system is the best means of achieving our genuine concerns about equality. If we can achieve more desirable policies to improve distributive justice like a universal basic income and a negative income tax, then we should definitely not try and engineer equality at the level of the university. In a system with overall distributive justice, differences in education are down to different choices and shouldn't trouble us.

In sum, we can conclude that though neither of the main libertarian arguments against affirmative action hold, it may nevertheless be an undesirable scheme because it actually hurts those it intends to help. But a diverse system is surely preferable to a one-size-fits-all set of admissions policy, and that suggests we'd want to leave universities to decide whether or not they implement positive discrimination for themselves.

Adam Smith and distributive justice

Many libertarians are sceptical about the idea of social justice, citing Hayek's argument that social justice is a mirage. Indeed, recently David Friedman had a debate with Jason Brennan, John Tomasi and Matt Zwolinski of the Bleeding Heart Libertarians blog over whether the concept even had a clear meaning. My own view is that social justice is just justice writ large, with particular focus on distributive issues like equality, priority and sufficiency.

Many classical liberals were deeply interested in questions of distributive justice, including Adam Smith, who made his name as a moral philosopher, and often focused on the damaging effects mercantilist and other interventionist policies had on the worse-off. Barry Stocker has recently posted the text of three very interesting talks he gave in Istanbul on the subject of Smith and distributive justice. Stocker highlights the ways in which Smith laid the blame for unjust distributions of society's goods at the state's door:

The cause [of unjust distributions] is largely the activity of the state rather than the results of markets being left free of state legislation and government schemes. Smith sees injustice as resulting from collaboration between merchants in the same sector, but sees this as more the consequence of state intervention than of free commerce. The state enabling, encouraging and even requiring enterprises to form corporate bodies (such as local chambers of commerce in Britain) in the same sector is the biggest reason for merchants conspiring against the public. That is the source of the famous quotation about merchants conspiring against the public, though that quotation is often used to support demands for increased state regulation.

And Stocker also highlights how Smith's concern for social justice did not translate into calls for redistribution; he believed that a good overall institutional structure would generate desirable distributional outcomes:

One of the problems with Smith commentary is that admirable scholars and political theory thinkers, like Rasmussen and Fleischacker, who are disposed favourably to a theory of redistributive justice see it in those elements of Smith which express a wish for distributive justice. There is distributive justice in Smith in the sense that he favours the distribution that emerges from freedom in economic activities, and in the state measures he favours to benefit the poor rather than the rich. However, that is not the same as the kind of belief in a predetermined pattern of distribution of justice which Rawlsians, or egalitarian liberal favour, at the extreme a completely flat distribution as argued for by G.A. Cohen and which is in the basic assumptions of Habermas‘ thought on norms, ethics, and discourse.

Read the whole things: 1, 2, 3.

Globalization drives cultural diversity

Donald Boudreaux recently reposted this 2010 essay on the impact of globalization on culture. Globalization is not about 'just stuff', he says, it's about increasing diversity by allowing different parts of different cultures to mix:

A century ago, there were no internationally franchised restaurants in Paris, France or, for that matter, in Paris, Texas. A century ago, residents of neither Omaha, Nebraska nor Birmingham, England could find sushi restaurants near their homes; today, sushi restaurants are all over the Western world. A century ago, blue jeans were not the international fashion that they are today. A century ago, the typical man's business suit worn by New York lawyers and London bankers was not widely worn in Africa and Asia, as it is today. In many ways, global commerce has indeed made the world more homogeneous.

But look more closely. While the differences between Paris, France and Paris, Texas are fewer than they were in the past, the cultural richness of each of these places today is far greater than it was just a few years ago. For a resident of Paris, Texas, circa 2010, the richness of the cultural smorgasbord available to him or her right at home is vast. A Texan can stay in town and dine on Vietnamese, Italian, or Greek food—or on barbeque. A Texan can listen to German symphonic music or medieval chants or Irish dance music or Edith Piaf—or country and western. A Texan can buy French neckties, English raincoats, and Italian scarves—and cowboy boots. Likewise a Parisian can choose croissants or New-York-style bagels. A mere century ago—even thirty years ago—the cultural diversity of both places was much less than it is today.

It's easy to be annoyed at the 'touristification' of a place like Thailand, but what that really means is more people get to experience somewhere they would only be able to imagine visiting fifty years ago. Perhaps it's no coincidence that this complaint usually comes from the people who can most easily afford foreign holidays and expensive exotic meals in their home cities. I'm tempted to say that they should check their privilege.

Boudreaux's piece is worth reading in full.

Don't hate the players, hate the game

I usually agree with Mark Littlewood, Director-General of the IEA, so I was surprised by his piece in the Mail on Sunday this weekend. Mark proposes a public register of everyone claiming benefits of any kind – pensions, disability living allowance, jobseeker’s allowance, and so on. This strikes me as a very bad idea indeed.

Mark’s aim is to increase public awareness of benefits claimants who are receiving much more in benefits than most people would think reasonable. This, he hopes, will increase the public’s appetite for welfare cuts. Actually, I think people overestimate how much money individual people on benefits get, but the proposals are undesirable for other reasons.

Mark says that “This wouldn’t be a matter of ‘naming and shaming’ anyone. After all, if you are legally entitled to a particular benefit, what is there to be ashamed about? Anyone ashamed to claim money from the State maybe shouldn’t be claiming it.”

In my experience, most unemployed people are profoundly ashamed of being unemployed. Removing their privacy, exposing them to gossiping neighbours and their children to bullying classmates, will just make that even worse.

And Jobseeker’s Allowance only accounts for a small proportion of the welfare budget. These proposals would also include people on disability benefits for socially stigmatized mental illnesses and physical disabilities that they would like to keep private.

Mark says that Britons “are far too reasonable to start taking up pitchforks and burning torches and assaulting imagined benefit cheats.” I am less sure. This is, of course, the same country that saw a paediatrician being hounded by vandals who confused the word “paediatrician” with “paedophile”.

These proposals would humiliate people on benefits and rob them of their privacy. They don’t deserve it. Many (probably most) of them are dependent on welfare because of the state itself, and it is senseless to make their lives even more difficult instead of tackling the real causes of their poverty.

If you think that unemployment is largely caused by government mismanagement of the economy, it makes no sense to humiliate people for being out of work. If you think that government welfare has crowded out private charity, you shouldn’t blame people forced to rely on government disability benefits. If you blame planning regulations for the high cost of housing, you should focus on those regulations before you cut off the money that mitigates the problem for a few poor people.

I wish the only problem today was the government’s unwillingness to cut spending. In fact, that spending usually exists to relieve much bigger problems that can’t be found on the Treasury balance sheet. Often, those problems are state-made.

To me, this is one of the key messages that ‘bleeding heart’ libertarians need to get across to other free marketeers. Cutting back the state is a bit like a game of Jenga – if you blithely pull away the supports that people rely on before you take away the causes of that reliance, you’ll only end up making things worse.


The rise and fall of the Gold Standard

George Selgin, prominent monetary theorist and blogger at, who recently gave an excellent talk at the ASI on "good deflation", wrote a history of the gold standard in the USA, explaining that there is no one narrative or theme throughout the history, with the fortunes of gold rising and falling with the times. While he pokes holes in some of the common garden arguments against a return to gold he also has his own reasons for distrusting a new regime founded on the yellow metal:

The claim that the real price of gold has become too volatile to allow that metal to be relied upon as a standard, for example, overlooks the extent to which gold’s price depends on the demand for private gold hoards, which has become both very great and very volatile precisely because of the uncertainty that fiat money regimes have inspired. The claim also overlooks the tendency for a metal’s price to become more stable as it becomes more widely adopted as a monetary standard.

Nor is it the case that there is not enough gold in the United States to support a new gold standard. According to Lawrence White, the Treasury’s gold stock, assuming that it is indeed what the Treasury itself claims, would at an official gold price of $1,600 per troy ounce be worth almost 20 percent of 2012 M1, making for “a more than healthy reserve ratio by historical standards.”

There are, however, some more compelling reasons for doubting that a return to gold would prove worthwhile. One is the prospect that any restoration of the convertibility of dollars into gold might be so disruptive that the short-run costs of the reform would outweigh any long-run gains it might bring. A second compelling reason has to do with the specific disadvantage of a unilateral return to gold. Here, once again, it must be recalled that the historical gold standard that is remembered as having performed so well was an international gold standard, and that the advantages in question were to a large extent advantages due to belonging to a very large monetary network.

Finally and perhaps most importantly, it is more doubtful than ever before that any government-sponsored and -administered gold standard would be sufficiently credible to either be spared from or to withstand redemption runs.

Read the whole thing.

Health inequality isn't all caused by income inequality

Chris Snowden has picked up on something that has long been a bugbear of mine. I shouted about it back when the Marmot Review on health inequality came out. It simple isn't true that all health inequality is as a result of income inequality: but that was the stance that the Review took.

Poor health will likely lead to low incomes, for example (reverse causation)

Absolutely: there are two effects going on. Getting some ghastly chronic disease in your 40s is obviously going to make you poorer in your 60s than if uyou'd been able to continue your meteoric rise up the career ladder to glory and a CEO's paycheque. I have no doubt that income inequality leads to some health inequality: I'd be surprised to find rich children suffering from vitamin deficiencies for example (assuming that Mother doesn't try all of the Mail's diet advice on her anklebiters) for example. But it's also true that health inequality leads to income inequality.

There's another effect going on as well. We're annually reminded (when the figures come out) about the geography of health inequality too. Men in Manchester or Glasgow die younger than those in Eastbourne for example. But again we're not being told a very important part of the story: people do move around you know. So it isn't true that someone born in Glasgow is destined for an early death: rather, it's those who don't climb the ladder up out of the slums who are. And the reason that lives are so long in Eastbourne or other retirement hotspots is that people only move to them when they are indeed retiring. And age expectations at 65 are very much higher than expected life span at birth. Simply because you've already survived, by definition, all of the things that were going to kill you before you got to 65.

Along with Snowden I tend to think that there are certain sets of statistics that are deliberately misrepresented in order to lead to a desired political conclusion. And those on health and age at death inequality are two sets of them.

The costs of regulation and why we're not creating enough jobs

We all know that regulation has benefits. We all also know that regulation has costs. The usual political mantra is that the costs are minimal while the benefits are huge. That may not in fact be true:

Regulation’s overall effect on output’s growth rate is negative and substantial. Federal regulations added over the past fifty years have reduced real output growth by about two percentage points on average over the period 1949-2005. That reduction in the growth rate has led to an accumulated reduction in GDP of about $38.8 trillion as of the end of 2011. That is, GDP at the end of 2011 would have been $53.9 trillion instead of $15.1 trillion if regulation had remained at its 1949 level.

It's worth thinking about that for a moment. Each individual American, the society as a whole, would be three times richer than they are if there had not been that explosion of regulation of the economy since WWII. That sort of increase in wealth buys quite a lot of people harmed by the lack of regulation.

But it's possible to use this to explain a disturbing feature of today's problems as well:

Why the change? The arguments rooted in technological developments sound like this: "Technologies like the Web, artificial intelligence, big data, and improved analytics—all made possible by the ever increasing availability of cheap computing power and storage capacity—are automating many routine tasks. Countless traditional white-collar jobs, such as many in the post office and in customer service, have disappeared. W. Brian Arthur, a visiting researcher at the Xerox Palo Alto Research Center’s intelligence systems lab and a former economics professor at Stanford University, calls it the “autonomous economy.” It’s far more subtle than the idea of robots and automation doing human jobs, he says: it involves “digital processes talking to other digital processes and creating new processes,” enabling us to do many things with fewer people and making yet other human jobs obsolete.

It has always been true that technological advance destroys jobs. But it has also always been true that technological advance creates other jobs as well. There's a worry that this isn't happening in the current economy. And that first paper gives us a clue as to why. There's simply too much regulation. If the economy were three times larger than it currently is then I do rather doubt that there would be much unemployment. And even if we take their numbers as being a tad fantastical, their basic point is obviously sound. Regulation restricts economic growth. It's economic growth that produces jobs. We're not creating enough jobs thus we've not got enough growth (and do recall, growth must be above labour productivity growth for there to be any expansion in employment) and over regulation is at least a part of that problem.

So let's hang the bureaucrats in order to get the unemployed back to work.

Check your fallacies

The latest game on the left is called "check your privilege," and if you make any point about others perhaps less advantaged than you, you are given to understand that you cannot really comment objectively, given your advantage.

It’s quite an old game.  Marxists used to call it "sociology of knowledge", but the rules were similar.  All of your opinions were alleged to be only the product of your class interest, and could therefore be discounted.  If you advocated market economics and classical liberalism, for example, this was simply an expression of your class interest as a member of the bourgeoisie.  It has the advantage that the intellectual content of your views can be ignored.  Opponents do not have to argue with what you say; since it represents only your class interest it can be ignored.  There is an exception.  One group is sufficiently detached from the class system that their views have objective import.  These are the Marxist intellectuals, of course.

The fallacies in "check your privilege" are straightforward and easy to identify, though Herbert Marcuse (remember him?) would no doubt have dismissed them as part of "bourgeois logic."  First is the argumentum ad hominem In which what is said is discounted, not because of any flaw or fault in its argument, but because of something pertaining to the arguer.  It is not the substance or sense of what is said that is being criticized, but the status of the person putting it forward.  The fallacy lies in the fact that the argument itself is not addressed, but irrelevant material is considered in its place.

The second fallacy is the genetic fallacy.  Despite the name this has nothing to do with Darwin or Mendel, but involves a dislike of where an argument comes from.  People are less inclined to accept views from those they dislike, whatever the merits of the actual views.  The mistake is to suppose that the source of an argument affects its validity.  A common meme is to assume that eventually someone will associate one side of an argument with Adolf Hitler, but it is still committed if you think that the views of rich white males can be discounted because of the three categories of those holding them. 

Other fallacies are touched on, but all belong to the category of informal fallacies of relevance (intrusion), and represent considering that qualities pertaining to the arguer somehow undermine and diminish the argument.  They don't.  In its latest form it is simply an anti-intellectual way of doing down what the other side is saying without facing the difficulty of considering their argument.

A Hayekian argument for equality of wealth

According to Friedrich Hayek, star of interwar economics, and recently star of two excellent rap battles with his contemporary John Maynard Keynes, the type of equality that matters is equality before the law. Equality before the law—and not equality of opportunity, outcome or anything else—is absolutely central to his political philosophy. But I think his economic work, particularly his crowning achievement, "The Use of Knowledge in Society" (cited a mere 9,496 times), points to the importance of a different kind of equality.

In "The Use of Knowledge in Society" and in earlier work in the socialist calculation debate Hayek shows that scientific knowledge isn't the totality of knowledge in society. In fact, a much bigger body of knowledge—information about peculiarities of time and place and preferences—is dispersed extremely widely across individuals. The only effective way of using this knowledge is a market system. Market participants act on the information embedded in different prices, and in doing so send yet more information back in other prices. But each participants learns only what she needs to know.

The market system Hayek envisions is a great practical means of organising society to achieve high social welfare. The flaw with this system is that participants with more money get to send stronger signals than others.

Markets measure how intensely we want things much more accurately than most democratic systems, because individuals have to bid against others for desirable goods or services. But this breaks down if individuals lack equal wealth. Ten pounds spent by a pauper is likely to represent a much more intense preference than that same £10 spent by a billionaire, a millionaire, or even the average middle class homeowner. However, producers will treat these £10s the same, and the economy will be skewed towards satisfying wealthier people's preferences.

Actually, it's not as simple as that. One feature of the market system is that people get rewarded with more money income (and potentially wealth) if they choose less leisure, or a riskier or less satisfying job. These sorts of inequalities, even if they produce wealth inequalities, would not subvert the system. These individuals have paid for their higher wealth with lower utility in work—and extra wealth merely evens out the overall extent to which the economic system is tilted to their advantage.

But endowments of talent or wealth through better upbringing, genetic advantage or inheritance do subvert the system, and undermine Hayek's argument for the efficiency and rationality of the market order by counting some people's preferences as more than one.

Now, by no means am I saying it's easy to disentangle these "good" sources of unequal wealth from the "bad" sources, or even that we ought to try to do so, and then even out endowments. But I do think that the effect inequality of wealth has on the market functions as a strong—and Hayekian—reason to desire a flatter distribution.